The Gateway: Reloaded

Just recently, the “gateway theory” has been doing the rounds yet again. I stumbled across this paper that claims the gateway theory is in fact, real and undeniable.

As usual, the researchers are making wild claims about cause and effect, but there’s one key problem with this paper, in that it is the outcome of focus groups:

It is a form of qualitative research consisting of interviews in which a group of people are asked about their perceptions, opinions, beliefs, and attitudes towards a product, service, concept, advertisement, idea, or packaging.


Let’s look at the focus group paper. I’ve already had a few thoughts on this on Twitter, where I picked out some of the more, er, ludicrous aspects.  The title on its own (Adolescents and young adults’ perceptions of electronic cigarettes as a gateway to smoking: a qualitative study in Switzerland) immediately puts me in the mind of “the authors had a preconceived idea, and needed to prove it” type thinking.

Don’t get me wrong, that’s how most science is usually done. Come up with a theory or hypothesis, then seek to prove (or disprove) it via experimentation. The difference with tobacco control ‘research’ is that the experimentation is often skewed (deliberately in some cases) to fit the theory.

From the abstract of this paper, it is abundantly clear that the authors intend to “prove” that the gateway theory is real, and is a major concern that, of course, requires immediate action:

Electronic cigarettes (ECs) acting as a gateway to smoking traditional cigarettes (TCs) is a growing public health concern of EC use among youths. To gather the opinions and perceptions of adolescents and young adults (AYAs) on whether and how EC can act as a gateway to smoking TC among youths. A qualitative method included 42 AYAs.

Right from the start, this paper is nothing more than a glorified opinion piece. What’s the saying? the plural of anecdote is not data?

We conducted a qualitative study on EC use in the French-speaking region of Switzerland, including 42 AYAs (19 females) aged 16–26 years. Eight focus groups (FG) were led: 4 with current EC users, 2 with TC users, 1 with non-users, and 1 with mixed consumption types. Group size ranged from 2 to 8 participants. Groups were stratified by consumption type to determine if there are any differences by user types.

The authors make a point of highlighting how many females were involved in this, for reasons known only to the authors. Perhaps they feel that women can provide better quality information than men? If that’s the case, why not a larger number of women? Either way, the participants were recruited through online ads (on ‘adolescent oriented websites) and offline (universities, clinics etc). Each participant was paid ~40$ (by a department store gift card) to thank them for their time.

Now, my first thought there is, people will do almost anything to get something for nothing. It’s human nature, we all, at some point, want to extract the most reward, for the least of effort. This isn’t necessarily a flaw in the paper, as offering an incentive can and does, encourage folk that wouldn’t normally get involved.

A quick breakdown of the 42 participants is needed at this point:

  • 31 were currently using or had ever used, e-cigarettes
  • 37 were currently smoking or had ever smoked cigarettes
  • 13 were currently dual-users
  • 4 had never used either

Considering that 13 were currently dual-users, and some of the e-cig and cigarette groups had “ever used” (which I read to say, had used in the past but don’t now) the balance of opinion is somewhat skewed in one direction. Very little information is contained in the paper about their habit. Are the users “regular” or “occasional”. It isn’t clear, all we get from this paper is that breakdown – which clearly doesn’t add up to 42. I’d probably flag that as a flaw as users tend to have a biased perception of the product they are using, and in the case of smoking versus vaping, that bias is even stronger. The only “balance” is provided by a tiny group of 4 individuals who would have the “outsiders perspective”.

Overall, participants identified a significant risk of EC easing the access to TC and acting as a gateway to TC use among adolescents: ‘I believe that it’s an open door to start smoking cigarettes, clearly’ (23-year-old male TC and EC user).

Given the breakdown of participants, along with the current general perceptions of vaping, this comment comes as no surprise to me at all.

The main reason being because ECs have the effect of erasing boundaries: ‘There used to be a barrier that said either you’re a smoker or a non-smoker, now I can smoke without smoking.’ (20-year-old male TC and EC user).

Well, no. Vaping is markedly different from smoking. It does simulate smoking to a certain degree, and for many, it is a replacement for smoking. However, in the eyes of some, if you vape you are a non-smoker. Y’see, this is what happens when the wrong language is used, and there is a deliberate conflation of smoking & vaping. Many will believe that one is the same as the other, regardless of how different they actually are.

Most of them agreed that while ECs can be fine for smokers, they can be a problem for nonsmokers: ‘I mean it’s better if the smokers use them [ECs], not those who have never touched [cigarettes]. That’s where the problem is.’ (24-year-old male TC and EC user);

I found it interesting that the highlighted quotes in the main “results” section are predominantly dual-users. It’s almost as if the authors identified comments that reflected their views to include, isn’t it?

The rest of the results section is further broken into specific sub-topics:

  • Preparing the ground for smoking
  • A facilitated transition to smoking
  • A reminder of the pleasures of smoking
  • An attractive and fashionable object
  • Its perception as a harmless toy
  • Other opinions about a gateway effect

Yes really. I wonder what exactly was included in the study preamble given to the participants, how much the press has influenced the participants’ opinions – regardless of their “use status”, and finally, the environment that the focus group discussion took place. It may indeed be the participants’ opinions, but I do wonder exactly how those opinions are formed. I highly doubt it is through use of a product only.

Out of all the opinions cited in this paper, the vast majority of them claim that ecigs are a gateway to cigarettes. There’s a small minority, the exact figure is unknown as it isn’t included in the paper – not surprising – nor are any other figures included to support the other statements cited.

Only a minority of participants did not perceive EC as a risk for a gateway to TC for two reasons. First EC could have the reverse effect of keeping adolescents away fromTC: ‘I think it’s not that bad that the tendency goes towards ECs for the young ones because if it wasn’t for the EC they would all turn to tobacco.’ (20-year-old male EC and TC user); because of the differences in taste:‘Youths now discover the EC, there are lots of flavors, and then they’ll say to themselves ‘why would I turn to tobacco’. [. . .] They will get used to these flavors and will find tobacco cigarettes disgusting, which is good.’ (20-year-old male EC and TC user). Second, some believed that no one would want to start ECs if not smoking TCs in the first place: ‘Why would one want to start using an EC if one doesn’t smoke in the first place?’ (16-year-old female TC-only user).

That’s it. The only statements contradicting the authors’ preconceived notion of a gateway theory. It is highly possible that a ‘gateway’ does exist for a small number of people, as of yet no study has accurately detected such a phenomenon.

Naturally, the authors highlight one flaw in their “novel research” in that the participants didn’t differentiate between first-generation and later e-cigs. They don’t point out any of the flaws that I noticed (natch), but go on to make recommendations anyway:

several recommendations stem from this study. Health professionals should screen adolescents (even very young ones) for EC use and inform consumers of the potential gateway effect.

Assuming that use of product A leads to use of product B, which hasn’t been conclusively proven in either direction.

The potential gateway effect of ECs should be acknowledged to break up with the harmless perception of ECs many might have; and preventive measures should be taken in this regard for young non-smokers and former smokers.

“We can’t prove a gateway, but let’s tell folk that ecigs aren’t harmless anyway”

From a public health perspective, there is an urge for better preventive and regulatory policies directed at protecting adolescents and children who have never smoked and support those who have quit smoking.

Let’s impose more regulations to protect The Children™.

It’s like the authors’ already had those recommendations in mind when they postulated The Question, isn’t it? (The answer to The Question is 42. Always 42.)

Thing is, this paper will carry a reasonable amount of weight, mostly because there aren’t any tenuous ties to “industry”, unlike a certain other paper that came up with the 95% figure.

Here we have two papers, one that has a vaguely tenuous link via a very circuitous route to “industry” and came up with the general conclusion that e-cigs are “95% safer than cigarettes”, which prompted a slew of articles claiming “industry influence” including an anonymous editorial in The Lancet and an article penned by McKee The BMJ.

On the flip side, we now have this paper that “proves” the gateway theory (based solely on nothing more than opinions of 42 young people) with seemingly no ties to “industry” tenuous or otherwise.

Which do you think will have the greater influence?

(image credit SergeBertasiusPhotography/